..
!
!
I
Reprinted from The Florida Entomologist, Volume 68, Number 4, December, 1985.
PREDICTION OF LARVAL INFESTATION I N
PASTURE GRASSES BY SPODOPTERA FRUGIPERDA
(LEPIDOPTERA NOCTUIDAE)
FROM ESTIMATES O F ADULT ABUNDANCE
J. F. SILVAIN& J. TI-A-HING
Laboratoire d'Entomologie appliquée
Institut Français de Recherche Scientifique pour le
Développement en Coopération
Centre ORSTOM de Cayenne
B.P. 165, 97323 Cayenne Cedex, French Guiana
AsSmAm
The .seasonal fluctuations of adult Spodopteru f m g i p e r d a (J. E. Smith) populations
were monitored for a 75-week period in pastures on a cattle farm in French Guiana.
Adults were monitored using blacklight traps and sticky traps baited with (Z)-g-dodecen-1-01 acetate, an attractant for S. f m g i p e r d u males. Larval populations were sampled
weekly using sweep nets. Significantcorrelations were obtained between adult captures
for both trapping methods used and relative larval abundance. The highest correlation
coefficient was obtained between weekly captures of moths in sticky traps and relative
larval abundance one week later. The highest populations of moths and larvae were
observed during the rainy seasons, and the lowest during the dry seasons. Results
obtained in these experiments demonstrate that it is possible to use sticky traps baited
with (Z)-9-dodecen-l-ol acetate in a system to warn growers of possible significant
infestations of S. N g i p e r d a larvae in pastures in French Guiana.
RESUMEN
Las fluctuaciones estacionales de las poblaciones adultas de Spodopteru fiugiperda
fueron seguidas durante 75 semanas en las pastos de una finca de cría de ganado en
Guayana Francesa. Los adultos fueron seguidos por medio de trampas de luz negra y
trampas pegajosas cebadas con (Z)-9-doaecen-l-olacetato, un atrayente para los machos
de S. fmgiperda. Se recogieron muestras de poblaciones larvales cacla semana,
empleando el método de la red de siega. Correlaciones significativas fueron obtenidas
entre las capturas de adultos por ambas trampas y la abundancia relativa de larvas. El
coeficiente de correlación más elevado fue ,obtenido al comparar las capturas semanales
de los adultos con trampas pegajosas y la abundancia relativa de larvas una semana más
F
Silvain & Ti-A-Hing: Pheromones to Monitor Spodoptera 687
tarde. Las mayores poblaciones de adultos y de larvas fueron observadas durante la
estación de lluvias, y las más bajas durante el período de sequía. Los resultados obtenidos en estos experimentos demuestran que se pueden utilizar trampas pegajosas
cebadas con (Z)-9-dodecen-l-ol acetato en un sistema destinado a caucionar a los
ganaderos sobre posible infestación de S. frugiperda en pastizales en la Guayana Francesa.
. .
Efforts to advance agricultural development in French Guiana have given priority
to the establishment of intensive cattle farming on improved pastures and to rice production. Spodoptera f m g i p e r d a (J. E. Smith) is one of the most important insect pests of
forage grasses, rice and soybeans in French Guiana. During certain years forage grasses, particularly Digitaria swazilandensis Stent which is most widely used in French
Guiana sustain extensive damage due to population explosions of S. f q i p e r d a larvae
(Silvain & Thiberville 1984a). These infestations require insecticide (Deltamethrine)
applications. However, without regular monitoring of pastures, these applications often
are carried out when populations have already reached the last larval instar or even the
pupal stage, and their impact is minimal.
Although much research has been carried out on monitoring techniques for S.
frugiperda in North America (Tingle & Mitchell 1975, 1977, 1979; Waddill et al 1982;
Starrat & McLeod 1982), on annual crops, little is known about seasonal abundance of
this pest in perennial crops such as forage grasses (Lynch et al 1980). In tropical areas
with continuous generations the prediction problems are different (Silvain & Thiberville
1984a). This paper reports experiments conducted from November 1981 to April 1983
to asses the possibility of using blacklight or pheromone traps to monitor S. frugiperda
moths and to predict subsequent larval outbreaks.
MATERIALS AND METHODS
The experiments were carried out on a cattle farm in Matoury (52" 20' W, 4" 51' N)
located 15 km from Cayenne, the capital city of French Guiana. This farm consists of
about 60 ha. of pastures divided into 19 fields covered by an imported creeping grass,
D. swazilandensis. The fields are grazed on a 45-day cycle and fertilized once a year.
To the south and the east of the farm remnents of the original rain forest remain, while
to the north and the west cultivated farmland is used for truck farming and sugar cane.
Meteorological information was obtained from instruments installed on the farm and
from the Meteorological Station at Rochambeau airport. There are two main seasons in
French Guiana: the rainy season from mid-November to mid-August and the dry season
"
from mid-August t o mid-November. There also may be a period of sunny weather in
February-March. The site we used is situated in a climatic zone characterized by: high
annual rainfall, with large variation from year to year; long pronounced dry season,
during which crops may suffer from lack of water; and very little fluctuation in temperature.
Three PheroconB 1C sticky traps baited with 25 mg of (2)-9-dodecen-1-01acetate in
ConrelB hollow fiber dispensers (50 hollow fiberddispenser) were located at the three
corners of an equilateral triangle, whose sides measured approximately 500 m. Considering the surface area of the farm, this represented one trap for 20 ha. The traps were
suspended one meter above the ground on a metal support and covered by a metal roof
to protect the traps from heavy rain. The traps were operated each week from Friday
morning through Monday morning. The sticky liners were changed weekly and the bait
every two weeks.
688
i
I
3
Florida Entomologist 68(4)
December, 1985
An Ellisco@blacklight trap (type PT 110103 PC) fitted with a 15 W G.E. F15T8BL
lamp was placed two meters from one of the pheromone traps in the southern angle of
the triangle. The blacklight trap was operated on Monday and Thursday nights of each
week. This method avoided possible interference between the two types of trap used.
Daily trapping, however desirable, was not feasible in this present setting. Cotton,
soaked with ethyl acetate, was placed in the bottom of the B.L. trap to kill the captured
insects.
To estimate the seasonal abundance of S. f m g i p e r d a larvae over the study area, 6
of the 19 fields were sampled each week on Tuesdays and Fridays between 8 and 11
a.m. The experimenter chose a random path through each field and sampled larvae by
taking 100 sweep-net samples (diameter of the sweep-net: 30 cm; 180” technique.)
RESULTSAND DISCUSSION
P
In 75 weeks, 4481 S. fmgiperda males were captured by pheromone traps (x: 6.6 -C
7.3 males/trap/night) as well as 23 females and 6 S. latifascia Walker males. During the
same period 464 S. fmgiperda moths were captured by the blacklight trap (x: 3.1 5.1
mothshight) of which 81.8% were females. Callahan and Chapin (unpublished data cited
iniMitchel1 1979) also observed a predominance of females in B.L. traps in Louisiana,
of’which ca. 55% have mated one or more times. Preliminary studies have indicated
that approximately 90% of females attracted to B.L. traps in French Guiana have mated
(Silvain 1982, unpublished data). A total of 9158 larvae of various instars were collected
with the sweep net (Z: 23.8 +. 71.8 larvaell00 sweeps).
The capture of moths (both sexes) by the blacklight trap and males by pheromone
traps show very similar patterns (Figure la), an observation which was confirmed by
a Spearmann’s rank correlation test (rs=0.66, t=7.51, p=O.Ol). These results are almost
identical to those we obtained from November 1979 until October 1981(Silvain 1984b).
Fluctuations in relative larval abundance (Fig. lb) correlated closely to adult abundance in pheromone (r,=0.70, t=8.51, p=O.Ol) or B.L. (rs=0.59, t=6.37, p=O.Ol) traps.
Comparisons of moth captures in pheromone and B.L. traps with relative larval abundance observed in the weeks preceeding or following adult catches (Fig. 2) show that
the highest correlation coefficients were obtained between moth catches during a given
week and relative larval abundance the following week.
During the period of observation, the monthly rainfall on the study site varied from
667 mm (May 1982) to 22 mm (September 1982). The mean monthly temperature was
26.4 & 0.3”C, with minimal and maximal temperature being 22.8 +. 0.6% and 30 -t 1”C,
respectively. The mean monthly relative humidity was 81.8 2.7% with 66.1 +. 5% and
97.5 k 1%as the minimum and the maximum daily values. The number of hours of
sunshine each month varied from 95.4 (March 1982) t o 270.4 (October 1982). A visual
comparison of adult (Fig. la) and larval (Fig. lb) abundance with patterns in seasonal
precipitation (Fig. IC) reveals the highest S. f m g i p e r d a populations during the rainy
season, and the lowest in the middle of the dry season (September-October). During
the 15 weeks when the weekly rainfall was less than 25”
(15th August-end of
November), average larval abundance was 3.9 & 15.4, compared with 29.7 f 80.6
during the remaining wet period. A significant correlation was observed between
weekly rainfall and larval abundance 3 weeks later (p=0.05) and 4 weeks later (p=O.Ol),
as well as between rainfall and males caught in pheromone traps 3 weeks later (p=0.05).
A direct effect of rainfall on larval growth is unlikely; however, vegetative growth of
the pasture grass is stimulated by rainfall, making the grass more attractive to ovipositing moths. This would tend to explain the increased abundance of adult moths recorded
at approximately 3-week intervals.
The results of these experiments demonstrate that adult trapping may be used to
*
i
*
I
Silvain & Ti-A-Hing: Pheromones to Monitor Spodoptera 689
1981
1982
FIG.lC
z70i
210
1981
FIG.lb
301
Fig. la. Spodoptera f h q i p e r d a moths captured per night in light trap (-1
and in
pheromone traps (---). Matoury, F. G., 1981-1983. Values for the BL trap include both
males and females; values shown for the pheromone traps are males only. Fig. lb. Sp.
f m g i p e r d a larvae per 100 sweeps on Digitaria swazilandensis pastures. Fig. IC.Rainfall in Matoury, F.G.
690
Florida Entomologist 68(4)
December, 1985
0.8
0.7,
I-
z
o
w 0.6,
LL
LL
w
0.5'
O
u
-I5
0.4'
O
0.3.
w
5
a
0.2-
o
0.1-
-5
-4
-3
-i -i
o
+i
+i +i +4
+ri,
WEEKS BEFORE & A F T E R M O T H CATCHES
Fig. 2. Correlation coefficients between the number of moths captured in pheromone
traps (males only) and light trap (males and females) and the number of larvae per 100
sweeps O to 5 weeks before and after the moth catches.
predict subsequent larval infestation of S. frugiperda on improved pastures in Matoury.
The best correlations were obtained between males caught in pheromone traps and the
number of larvae collected by sweeping the following week. These results are very
similar to those reported for tobacco budworm, Heliothis wirescens (F.) on tobacco by
Tingle & Mitchell (1981). A practical warning system against S. frugiperda must provide
information on the presence of early larval instars if it is to allow timely application of
insecticides, as the later instars are of economicimportance (Martin et al. 1980).Approximately 85% of the S. fmgiperda larvae recovered using the sweep net sampling procedure described are in the Ist to 4th instars (Silvain, unpublished data). However, sweep
net sampling is prohibitively expensive. The results we present indicate that pheromone
traps may be a practical alternative, especially in the case of perennial crops in a tropical
area where climatic conditions favor the maintenance of breeding populations of S.
fmgiperda throughout most of the year.
ACKNOWLEDGEMENTS
I thank Dr. E. R. Mitchell (Insect Attractants Lab., Gainesville), Dr. J. McNeil
.
Silvain & Ti-A-Hing: Pheromones to Monitor Spodoptera 691
(Université Lavel, Québec, Canada) and Dr. M. S. Adams for their reviews and suggestions.
REFERENCESCITED
LYNCH,R. E., P. B. MARTIN, AND J. W. GARNER. 1980. Cultural manipulation of
coastal bermudagrass to avoid losses from the fall armyworm. Florida Ent. 63:
411-19.
MARTIN, P. B., B. R. WISEMAN,AND R. E. LYNCH. 1980. Action thresholds for fall
armyworms on grain sorghum and coastal bermudagrass. Florida Ent. 63: 375405.
MITCHELL,E. R. 1979. Monitoring adult populations of the fall armyworm. Florida
Ent. 6 2 91-8.
SILVAIN,
J. F. AND F. THIBERVILLE.1984a. The noctuids (Lepidoptera Noctuidae)
dangerous to food and industrial crops in French Guiana. Proc. Caribbean Food
Crops Society 19th Annual Meeting. Puerto Rico: 217-239.
SILVAIN,J. F. 198413. Premières observations sur l’écologiede Spodopteru frugiperda
(J.E. Smith) et M o c k latipes (Guenée), noctuelles déprédatrices des graminées
foumagéres en Guyane française. In “Les Colloques de l’INRA”, n”24: Prairies
guyanaises et élevage bovin: 243-272.
STARRAT,A. N. AND D. G. R. MCLEOD. 1982. Monitoring fall armyworm, Spodoptera fnigiperdu (Lepidoptera: Noctuidae), moth population in southwestern Ontario with sex pheromone traps. Canadian Ent. 114: 545-9.
TINGLE,F. c. AND E. R. MITCHELL. 1975. Capture of Spodopteru fnbgiperda and s.
exigua. in pheromone traps. J. Econ. Ent. 68: 613-15.
TINGLE,F. C. AND E. R. MITCHELL. 1977. Seasonal populations of armyworms and
loopers at Hastings, Florida. Florida Ent. 60: 115-22.
TINGLE,F. C. AND E. R. MITCHELL. 1979. Spodopteru frugiperdu: Factors affecting
pheromone trap catches in corn and peanuts. Environ. Ent. 8: 989-92.
TINGLE,F. C. AND E. R. MITCHELL. 1981. Relationships between pheromone trap
catches of male tobacco budworm, larval infestations, and damage levels in tobacco. J.Econ. Ent. 7 4 437-440.
WADDILL,v. H.,E. R. MITCHELL,AND W. H. DENTON,s. L. POE, AND D. J.
SCHUSTER. 1982. Seasonal abundance .of the fall armyworm and velvetbean
caterpillar (Lepidoptera: Noctuidae) at four locations in Florida. Florida Ent. 65:
350-4.
i